Method for frequency regulation of air volume and temperature control of annealing furnace

Abstract

The application discloses an annealing furnace air volume frequency regulation and temperature control method, relates to the float glass technology field, and comprises the following steps: collecting operation basic data such as space temperature of each area of an annealing furnace and glass ribbon surface measuring point temperature, constructing a thermal inertia stress prediction model based on heat transfer characteristics and glass stress generation mechanism, predicting the thermal inertia lag amount of the furnace body and the potential stress risk of the glass ribbon through the model, combining temperature deviation and the prediction result to generate a basic air volume regulation signal, transmitting the signal to a frequency converter controller of a fan to complete preliminary regulation, collecting verification data after the preliminary regulation, judging the regulation effect, executing air volume maintenance or parameter correction operation according to the result, and through closed loop circulation until the temperature of each area is stable and the stress risk of the glass ribbon is in a preset range. The application improves the timeliness and accuracy of regulation, effectively controls the residual stress of the glass ribbon, and guarantees the stability of annealing quality.

Description

Technical Field

[0001] This invention relates to the field of float glass technology, and in particular to a method for frequency conversion regulation and temperature control of annealing furnace air volume. Background Technology

[0002] In the production of glass products such as float glass and photovoltaic glass, the annealing furnace is a key piece of equipment for eliminating residual stress in the glass and ensuring the mechanical strength and thermal stability of the glass products. The temperature control and air volume regulation of the annealing furnace are directly related to the quality of glass annealing. Among them, variable frequency air volume regulation technology has become the mainstream control method in the industry due to its advantages of energy saving and flexible adjustment.

[0003] In existing technologies, frequency conversion regulation of annealing kiln airflow is typically based on a DCS control system or a PLC controller. This involves collecting data such as the ambient temperature of each zone within the annealing kiln and the surface temperature of the glass ribbon, comparing this data with a preset target temperature to obtain the temperature deviation, and then generating an airflow regulation signal which is transmitted to the frequency converter of the fan. Airflow control is achieved by adjusting the fan speed. The core logic of this type of regulation is passive correction based on real-time temperature deviation, which can maintain kiln temperature stability to a certain extent.

[0004] However, annealing furnaces, as large-scale thermal equipment, exhibit significant thermal inertia hysteresis, and the generation of residual stress in the glass ribbon has a complex relationship with temperature gradient and cooling rate. Existing adjustment methods rely solely on real-time temperature deviations to adjust airflow, failing to consider the hysteresis effect caused by the furnace's thermal inertia and unable to predict potential stress risks in the glass ribbon under current operating conditions. This results in airflow adjustments often lagging behind temperature changes and stress evolution within the furnace. This deficiency directly leads to problems such as temperature overshoot and excessive residual stress in the glass ribbon during actual production, affecting the stability of glass annealing quality. Summary of the Invention

[0005] To address the technical problems existing in the prior art, this invention provides a method for frequency conversion regulation and temperature control of air volume in annealing kilns.

[0006] The technical solution adopted in this invention is:

[0007] This application provides a method for frequency conversion regulation and temperature control of air volume in an annealing kiln, including the following steps:

[0008] Step 1: Collect basic operating data of the annealing furnace, and construct a thermal inertial stress prediction model based on the correlation mechanism between the heat transfer characteristics of the annealing furnace and the stress generation of glass.

[0009] Step 2: Using the aforementioned prediction model and combined with operational data, predict the thermal inertia hysteresis of the annealing furnace and the potential stress risk of the glass strip.

[0010] Step 3: Based on the predicted thermal inertia hysteresis and temperature deviation in the basic operating data, a basic air volume adjustment signal is generated and transmitted to the frequency converter of the fan to complete the initial air volume adjustment.

[0011] Step 4: After the initial airflow adjustment, collect verification data and judge the adjustment effect based on the prediction model and verification data.

[0012] Step 5: Based on the judgment result, perform air volume maintenance or parameter correction operations, and repeat steps 4-5 until the temperature of each zone of the annealing furnace is stable and the stress risk of the glass strip is within the preset range.

[0013] Preferably, the basic data collected for the operation of the annealing kiln includes: collecting the spatial temperature of each zone of the annealing kiln, the temperature of multiple measuring points on the surface of the glass strip, the thickness of the glass strip, and the real-time operating parameters of the blower, wherein the multiple measuring points on the surface of the glass strip are distributed along its width direction and the length direction of the kiln zone.

[0014] Preferably, the construction of the thermal inertia stress prediction model in step 1 includes:

[0015] The transverse temperature gradient is calculated by measuring the temperature at different widths of the same cross section on the surface of the glass ribbon.

[0016] The longitudinal temperature gradient is calculated by measuring the temperature at the same width location in different kiln zones.

[0017] Based on the correlation mechanism between the heat transfer characteristics of each zone of the annealing furnace and the glass stress temperature gradient, a thermal inertia stress prediction model is constructed by integrating the transverse temperature gradient, the longitudinal temperature gradient and basic operational data.

[0018] Preferably, the predicted thermal inertia hysteresis of the annealing furnace and the potential stress risk of the glass ribbon include:

[0019] The collected glass strip thickness and the calculated transverse and longitudinal temperature gradients of the glass strip surface in each zone are input into the thermal inertia stress prediction model; the thermal inertia hysteresis of the annealing furnace under the current operating conditions and the potential stress risk level of the glass strip are output. The stress risk level is classified based on the correlation between the temperature gradient and the preset stress threshold.

[0020] Preferably, the step of generating a basic airflow adjustment signal based on the predicted thermal inertia hysteresis and temperature deviation in the basic operating data, and transmitting it to the frequency converter of the fan to complete the initial airflow adjustment, includes:

[0021] The collected spatial temperatures and glass surface temperatures of each zone are compared with the corresponding preset target temperatures to calculate the temperature deviation values ​​of each zone.

[0022] Based on the temperature deviation value and the thermal inertia hysteresis predicted in step 2, a collaborative calculation is performed to generate a basic air volume adjustment signal that is adapted to the current operating conditions.

[0023] The basic air volume adjustment signal is transmitted to the frequency converter control module of the fan in the corresponding area of ​​the annealing kiln, and the initial air volume output is achieved by adjusting the fan speed.

[0024] Preferably, the verification data collected after the initial airflow adjustment includes:

[0025] Within a preset period after initial airflow adjustment, real-time temperature response curves of each zone and temperature data from multiple measuring points on the glass strip surface are collected.

[0026] Repeat the calculation method of transverse and longitudinal temperature gradients in step 1, and update the current transverse and longitudinal temperature gradient data based on the temperature data of multiple measuring points on the newly collected glass strip surface.

[0027] The real-time temperature response curves of each zone and the updated horizontal and vertical temperature gradient data were used as validation data.

[0028] Preferably, judging the adjustment effect based on the prediction model and verification data includes:

[0029] Analyze the matching degree between the real-time temperature response curves of each zone and the thermal inertia hysteresis predicted in step 2, and determine whether the temperature response meets the preset adjustment expectations.

[0030] Determine whether the updated transverse and longitudinal temperature gradients on the glass strip surface are both within the corresponding preset allowable ranges.

[0031] If the temperature response meets the preset adjustment expectation and both the lateral and longitudinal temperature gradients are within the preset allowable range, the adjustment is deemed effective; otherwise, the adjustment is deemed ineffective.

[0032] Preferably, the parameter correction operation in step 5 includes:

[0033] If the adjustment is deemed ineffective, based on the thermal inertia-stress prediction model constructed in step 1, input the temperature response curve deviation and the temperature gradient data exceeding the threshold from the verification data;

[0034] By performing inverse model calculations, the air volume adjustment parameters are corrected, and an optimized adjustment signal is generated.

[0035] The optimized adjustment signal is fed back to the frequency converter control module to adjust the fan speed in the corresponding area and achieve air volume correction.

[0036] Preferably, the airflow maintenance in step 5 includes:

[0037] If the adjustment is deemed effective, maintain the current fan speed and air volume output, and continuously monitor the temperature changes in each zone of the annealing furnace and the temperature gradient of the glass ribbon until the temperature in each zone is stable within the target temperature range and the transverse and longitudinal temperature gradients of the glass ribbon remain within the preset range.

[0038] The beneficial effects of the present invention are at least one of the following:

[0039] By constructing a thermal inertia stress prediction model and predicting the thermal inertia hysteresis of the annealing furnace, the generation of the basic air volume adjustment signal can be fully adapted to the thermal inertia characteristics of the furnace body. This helps to reduce the hysteresis between air volume adjustment action and temperature change in the furnace, improve the timeliness and accuracy of air volume adjustment, and make the air volume adjustment and temperature control process of the annealing furnace more in line with the actual process requirements of glass annealing.

[0040] By using models to predict potential stress risks in glass strips and combining this with subsequent verification data, stress-related hazards can be avoided in advance, which helps improve the stability of glass annealing quality and reduce production risks caused by stress problems. This method makes the air volume adjustment and temperature control process of the annealing furnace more in line with the actual process requirements of glass annealing. Attached Figure Description

[0041] Figure 1 This is a schematic diagram of the method flow of the present invention. Detailed Implementation

[0042] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0043] In the production of float glass, photovoltaic glass, and other products, the temperature control and air volume regulation of the annealing furnace directly determine the residual stress level of the glass and the product qualification rate.

[0044] In existing technologies, frequency conversion regulation of annealing kiln air volume mostly relies on DCS control system or PLC controller. The core logic is to collect real-time temperature data and compare it with the preset target temperature to obtain the temperature deviation, and then generate an adjustment signal based on the deviation to control the fan speed.

[0045] However, as a large-scale thermal equipment, the annealing furnace has a complex structure and strong heat storage capacity, and has a significant thermal inertia lag characteristic. That is, after the air volume adjustment is completed, the temperature inside the furnace will not respond immediately, but there is a certain time delay. At the same time, the generation of residual stress in the glass ribbon is closely related to the temperature gradient and cooling rate. Existing technologies only focus on the passive correction of temperature deviation and cannot predict the stress risks that the glass ribbon may generate under the current operating conditions.

[0046] This often leads to problems such as temperature overshoot, asynchronous adjustment actions with kiln temperature changes, and excessive stress in the glass strip using existing control methods, affecting product quality stability. To address these issues, this embodiment provides a method for frequency conversion regulation and temperature control of the annealing kiln airflow, such as... Figure 1 As shown, it includes the following steps:

[0047] Step 1: Collect basic operating data of the annealing furnace, and construct a thermal inertial stress prediction model based on the correlation mechanism between the heat transfer characteristics of the annealing furnace and the stress generation of glass.

[0048] In one possible implementation, the collection of basic operating data of the annealing kiln includes: collecting the spatial temperature of each zone of the annealing kiln, the temperature of multiple measuring points on the surface of the glass strip, the thickness of the glass strip, and the real-time operating parameters of the blower, wherein the multiple measuring points on the surface of the glass strip are distributed along its width direction and the length direction of the kiln zone.

[0049] In one possible implementation, the step 1 of constructing the thermal inertia stress prediction model includes:

[0050] The transverse temperature gradient is calculated by measuring the temperature at different widths of the same cross section on the glass strip surface; the longitudinal temperature gradient is calculated by measuring the temperature at the same width in different kiln zones; based on the correlation mechanism between the heat transfer characteristics of each zone of the annealing kiln and the glass stress temperature gradient, a thermal inertia stress prediction model is constructed by integrating the transverse temperature gradient, the longitudinal temperature gradient, and basic operational data.

[0051] It should be noted that the basic operating data refers to the core data characterizing the operating status of the annealing furnace and the properties of the glass ribbon, including the temperature of each zone of the annealing furnace, the temperature of multiple measuring points on the surface of the glass ribbon, the thickness of the glass ribbon, and the real-time operating parameters of the fan.

[0052] The thermal inertia stress prediction model is a correlation model that integrates the heat transfer characteristics of the annealing furnace and the stress generation law of glass. It can output the thermal inertia hysteresis of the annealing furnace and the potential stress risk of the glass strip based on the basic operating data, and provide a predictive basis for air volume adjustment.

[0053] The transverse temperature gradient refers to the rate of temperature change along the width of the same cross section of the glass ribbon, reflecting the temperature uniformity along the width of the glass ribbon. The longitudinal temperature gradient refers to the rate of temperature change along the length of the kiln zone at the same width of the glass ribbon, reflecting the cooling rate of the glass ribbon during the annealing process.

[0054] Because the thermal inertia hysteresis characteristics of the annealing furnace and the stress generation of the glass strip are closely related to the temperature distribution, this step collects comprehensive basic operational data, explores the correlation between temperature gradient, glass thickness, fan parameters and thermal inertia and stress, and constructs a thermal inertia stress prediction model to achieve subsequent proactive prediction rather than passive correction.

[0055] During the implementation process, basic operational data is collected. The spatial temperature of each zone of the annealing kiln is collected by thermocouples installed in zones A, B, C, D, and F of the annealing kiln. 3-5 collection points are evenly arranged along the length of each zone to ensure full coverage of each zone.

[0056] The temperatures at multiple measuring points on the surface of the glass ribbon are collected by a radiation thermometer. Nine measuring points are evenly arranged along the width direction of the glass ribbon (including the edge, sub-edge and middle part), and one group is arranged at the entrance of zone A, the exit of zone B and the exit of zone C along the length direction of the kiln zone respectively, forming a three-dimensional measuring point network.

[0057] The thickness of the glass ribbon is collected by an on-line thickness gauge preset in the production line. The thickness gauge is installed at the transition roller table before the entrance of the annealing lehr, so as to obtain real-time glass ribbon thickness data. The real-time operating parameters of the fan are collected by the sensors matched with the fan, including the current rotating speed of the fan, output air volume and input power.

[0058] The calculation process of the transverse temperature gradient is as follows: select the temperatures at measuring points in the width direction of the same cross-section of the glass ribbon (e.g., the middle cross-section of zone B), and calculate the temperature change rate between adjacent measuring points. The formula is:

[0059]

[0060] wherein is the transverse temperature gradient, is the temperature of the i-th width measuring point, is the temperature of the (i+1)-th width measuring point, is the distance between adjacent width measuring points, and the maximum value of gradients of all adjacent measuring points is taken as the transverse temperature gradient of the cross-section.

[0061] The calculation process of the longitudinal temperature gradient is as follows: select the temperatures at measuring points in the length direction of the kiln zone at the same width position of the glass ribbon (e.g., the middle part of the glass ribbon), and calculate the temperature change rate between adjacent measuring points in the kiln zone. The formula is:

[0062]

[0063] wherein is the longitudinal temperature gradient, is the temperature of the j-th measuring point in the kiln zone, is the temperature of the (j+1)-th measuring point in the kiln zone, is the distance between adjacent measuring points in the kiln zone (i.e., the length of the corresponding kiln zone).

[0064] The construction process of the thermal inertia stress prediction model is as follows: based on the heat transfer characteristics of each zone of the annealing lehr, such as the difference between radiation heat transfer in zone A, hot air circulating heat transfer in zone B and forced convection heat transfer in zone C, and the correlation mechanism between glass stress and temperature gradient, the transverse temperature gradient and the longitudinal temperature gradient , glass ribbon thickness d, real-time rotating speed n of the fan, and the space temperature T of each zone are taken as input parameters of the model, and the thermal inertia hysteresis and the potential stress risk level R of the glass ribbon are taken as output parameters to construct a multiple linear correlation model. The model expression is:

[0065]

[0066] in These are the model coefficients. , The constant term is obtained by fitting historical operating data of the production line (covering measured data of temperature, gradient, and stress under different glass thicknesses and operating conditions).

[0067] Step 2: Using the aforementioned prediction model and combined with operational data, predict the thermal inertia hysteresis of the annealing furnace and the potential stress risk of the glass strip.

[0068] It should be noted that the thermal inertia hysteresis reflects the degree to which the thermal inertia of the kiln body affects the temperature response. Potential stress risk in the glass ribbon refers to the risk of residual stress exceeding the standard that may occur in the glass ribbon under current operating conditions. This risk is quantified by a stress risk level; a higher level indicates a greater likelihood of stress exceeding the standard.

[0069] Without anticipating thermal inertia hysteresis and stress risks, the generation of airflow adjustment signals lacks foresight, and adjustment actions often lag behind changes in kiln temperature and stress evolution. This step, based on the model built in step 1, inputs real-time collected operational data to obtain information on thermal inertia hysteresis and stress risks in advance. This allows subsequent airflow adjustments to adapt to kiln characteristics and glass stress generation trends in advance, improving the initiative and accuracy of the adjustment.

[0070] In the specific implementation process, the glass strip thickness d collected in real time in step 1, the calculated transverse temperature gradient G1 and longitudinal temperature gradient G2 of the glass strip surface in each zone, the real-time operating parameters of the fan (speed n, output air volume Q), and the spatial temperature T of each zone are simultaneously input into the thermal inertial stress prediction model.

[0071] Model Calculation: The model performs calculations based on the input real-time data using preset correlation formulas, namely the τ and R calculation formulas in step 1, and automatically completes the quantitative analysis and correlation derivation of the data.

[0072] After the model calculation is completed, two results are output: one is the thermal inertia hysteresis τ of the annealing furnace under the current operating conditions, and the other is the potential stress risk level R of the glass strip. The classification standard for stress risk level R is: R≤5 is low risk, 5<R≤8 is medium risk, and R>8 is high risk. This classification standard is based on the stress threshold requirements of glass products and is calibrated in combination with historical failure data of the production line.

[0073] Step 3: Based on the predicted thermal inertia hysteresis and temperature deviation in the basic operating data, a basic air volume adjustment signal is generated and transmitted to the frequency converter of the fan to complete the initial air volume adjustment.

[0074] In one possible implementation, the step of generating a basic airflow adjustment signal based on the predicted thermal inertia hysteresis and temperature deviation in the basic operating data, and transmitting it to the frequency converter of the fan to complete the initial airflow adjustment, includes:

[0075] The collected spatial temperatures and glass surface temperatures of each zone are compared with the corresponding preset target temperatures to calculate the temperature deviation values ​​of each zone.

[0076] Based on the temperature deviation value and the thermal inertia hysteresis predicted in step 2, a collaborative calculation is performed to generate a basic air volume adjustment signal that is adapted to the current operating conditions.

[0077] The basic air volume adjustment signal is transmitted to the frequency converter control module of the fan in the corresponding area of ​​the annealing kiln, and the initial air volume output is achieved by adjusting the fan speed.

[0078] It should be noted that the temperature deviation value refers to the difference between the actual collected spatial temperature and glass ribbon surface temperature in each zone of the annealing furnace and the corresponding preset target temperature. It is a core indicator reflecting the difference between the current temperature state and the ideal state. The basic air volume adjustment signal refers to the control signal generated based on the temperature deviation value and thermal inertia hysteresis. It is used to instruct the frequency converter of the fan to adjust the fan speed, thereby achieving preliminary optimization of the air volume. The frequency converter of the fan refers to the dedicated control equipment matched with the fans in each zone of the annealing furnace. It can receive the air volume adjustment signal and adjust the fan speed by changing the output frequency, thereby changing the fan output air volume.

[0079] This step combines the temperature deviation value with the predicted thermal inertia hysteresis to generate a basic air volume adjustment signal that is adapted to the thermal inertia characteristics of the kiln body. This allows the adjustment action to compensate for the hysteresis effect caused by thermal inertia in advance, thereby improving the accuracy of the initial air volume adjustment.

[0080] In the specific implementation process, the temperature deviation value is calculated as follows: the spatial temperature of each zone of the annealing furnace collected in step 1 is compared with the preset target temperature of the corresponding zone to obtain the spatial temperature deviation value of each zone. The formula is:

[0081]

[0082] The average temperature of multiple measuring points on the glass ribbon surface is compared with the preset target temperature of the glass ribbon surface to obtain the temperature deviation value of the glass ribbon surface. The formula is as follows:

[0083]

[0084] in, These are the spatial temperature deviation values ​​for each zone. The actual temperature of each zone. Preset target temperatures for each zone. This represents the surface temperature deviation of the glass ribbon. This represents the actual average temperature of the glass ribbon surface. Set a target temperature for the glass ribbon surface.

[0085] The process of generating the basic airflow adjustment signal is as follows: A collaborative calculation is performed based on the temperature deviation value and the thermal inertia hysteresis. The calculation formula is:

[0086]

[0087] Where S is the amplitude of the basic air volume adjustment signal. , , The adjustment coefficient is calibrated according to the heat transfer characteristics of different kiln areas.

[0088] The direction of the signal is determined by the sign of the temperature deviation value. or When the signal is positive, the direction is to increase the air volume; when it is negative, the direction is to decrease the air volume. Signal transmission and preliminary adjustment: The generated basic air volume adjustment signal is transmitted to the frequency converter controller of the fan in the corresponding area of ​​the annealing kiln via the communication bus. After receiving the signal, the frequency converter controller adjusts the output frequency according to the signal amplitude, thereby adjusting the fan speed to achieve preliminary air volume output; for example, when the signal amplitude S=10, the frequency converter controller adjusts the output frequency from 50Hz to 55Hz, the fan speed increases synchronously, and the output air volume increases.

[0089] This step generates an adjustment signal by coordinating the temperature deviation value and the thermal inertia hysteresis, enabling the adjustment action to adapt to the thermal inertia characteristics of the kiln body, compensating for the hysteresis effect in advance, avoiding the adjustment hysteresis problem caused by neglecting thermal inertia in existing technologies, and improving the accuracy of the initial air volume adjustment.

[0090] Step 4: After the initial airflow adjustment, collect verification data and judge the adjustment effect based on the prediction model and verification data.

[0091] In one possible implementation, collecting verification data after initial airflow adjustment includes:

[0092] Within a preset period after initial airflow adjustment, real-time temperature response curves of each zone and temperature data from multiple measuring points on the glass strip surface are collected.

[0093] Repeat the calculation method of transverse and longitudinal temperature gradients in step 1, and update the current transverse and longitudinal temperature gradient data based on the temperature data of multiple measuring points on the newly collected glass strip surface.

[0094] The real-time temperature response curves of each zone and the updated horizontal and vertical temperature gradient data were used as validation data.

[0095] In one possible implementation, determining the adjustment effect based on the prediction model and verification data includes:

[0096] Analyze the matching degree between the real-time temperature response curves of each zone and the thermal inertia hysteresis predicted in step 2, and determine whether the temperature response meets the preset adjustment expectation; determine whether the updated transverse temperature gradient and longitudinal temperature gradient of the glass strip surface are both within the corresponding preset allowable range; if the temperature response meets the preset adjustment expectation and the transverse and longitudinal temperature gradients are both within the preset allowable range, the adjustment is deemed effective; otherwise, the adjustment is deemed ineffective.

[0097] It should be noted that the verification data refers to the data collected after the initial air volume adjustment to verify the adjustment effect, including the real-time temperature response curves of each zone and the updated horizontal and vertical temperature gradient data, which can intuitively reflect the impact of the adjustment action on the temperature inside the kiln and the temperature distribution of the glass belt.

[0098] The adjustment effect refers to the degree to which the initial air volume adjustment action improves the temperature stability inside the kiln and the control of stress risk in the glass strip. It is determined by the matching degree between the temperature response curve and the thermal inertia hysteresis and whether the temperature gradient is within the preset allowable range.

[0099] After the initial airflow adjustment, it is necessary to verify whether the adjustment has achieved the expected effect through actual data to avoid adjustment failure due to operating condition fluctuations or model errors. This step collects verification data and combines it with the predictive model to judge the effect, providing a basis for subsequent maintenance or correction operations and forming a closed-loop control.

[0100] In the specific implementation process, within the prediction period after the initial air volume adjustment, the prediction period is set according to the thermal inertia lag, which is usually 1.5 times τ. For example, when τ=70s, the prediction period is 105s. Temperature data is continuously collected by thermocouples and radiation thermometers in each zone at a frequency of 1 time / 5s. The collected temperature data is arranged in chronological order to form the real-time temperature response curve of each zone.

[0101] During the prediction period, temperature data were simultaneously collected at 9 measuring points in the width direction and 3 measuring points in the length direction of the glass strip surface, with the acquisition frequency consistent with the temperature response curve.

[0102] Repeat the calculation method of the transverse and longitudinal temperature gradients in step 1, and recalculate the current transverse temperature gradient G1′ and longitudinal temperature gradient G2′ based on the newly collected temperature data of multiple measuring points on the glass strip surface.

[0103] The real-time temperature response curves of each zone and the updated lateral temperature gradient G1′ and longitudinal temperature gradient G2′ are integrated into verification data and transmitted to the control system.

[0104] Extract the settling time τ′ of the temperature response curve and compare τ′ with the predicted thermal inertia hysteresis τ from step 2. If |τ′-τ|≤τ×20%, the temperature response and thermal inertia hysteresis are considered to match; otherwise, they are considered to be mismatched. Temperature gradient range judgment: Determine whether the updated lateral temperature gradient G1′ is within the preset allowable range. The preset allowable range is set according to the glass thickness, such as ≤5°C / m for 5-10mm glass and ≤4°C / m for 10-20mm glass. Also determine whether the longitudinal temperature gradient G2′ is within the preset allowable range, such as ≤6°C / m for area A and ≤5°C / m for area B. If both are within the preset allowable range, the temperature gradient is considered to meet the requirements; otherwise, it is considered to be unacceptable. If the temperature response and thermal inertia hysteresis match and the temperature gradient meets the requirements, the adjustment is considered effective; otherwise, the adjustment is considered ineffective.

[0105] Step 5: Based on the judgment result, perform air volume maintenance or parameter correction operations, and repeat steps 4-5 until the temperature of each zone of the annealing furnace is stable and the stress risk of the glass strip is within the preset range.

[0106] In one possible implementation, the parameter correction operation in step 5 includes:

[0107] If the adjustment is deemed ineffective, based on the thermal inertia-stress prediction model constructed in step 1, input the temperature response curve deviation and the temperature gradient data exceeding the threshold from the verification data;

[0108] By performing inverse model calculations, the air volume adjustment parameters are corrected, and an optimized adjustment signal is generated.

[0109] The optimized adjustment signal is fed back to the frequency converter control module to adjust the fan speed in the corresponding area and achieve air volume correction.

[0110] In one possible implementation, maintaining the airflow in step 5 includes:

[0111] If the adjustment is deemed effective, maintain the current fan speed and air volume output, and continuously monitor the temperature changes in each zone of the annealing furnace and the temperature gradient of the glass ribbon until the temperature in each zone is stable within the target temperature range and the transverse and longitudinal temperature gradients of the glass ribbon remain within the preset range.

[0112] It should be noted that airflow maintenance operation refers to maintaining the current fan speed and airflow output when the adjustment is effective, continuously monitoring changes in operating conditions, and ensuring a stable annealing process. Parameter correction operation refers to correcting the airflow adjustment parameters based on verification data and predictive models when the adjustment is ineffective, generating an optimized adjustment signal, and further optimizing the airflow output. Closed-loop condition refers to the temperature in each zone of the annealing furnace stabilizing within the target temperature range, and the transverse and longitudinal temperature gradients of the glass ribbon remaining within the preset allowable range.

[0113] In the specific implementation process, the air volume maintenance operation is as follows: if step 4 determines that the adjustment is effective, the frequency converter of the fan maintains the current output frequency, and the fan maintains the current speed and air volume output state. The control system continuously monitors the temperature changes of each space in the annealing furnace, the surface temperature of the glass ribbon, and the temperature gradient, with a monitoring frequency of once every 10 seconds, and determines in real time whether the closed-loop conditions are met.

[0114] The parameter correction process is as follows: If step 4 determines that the adjustment is ineffective, the temperature response curve deviation (τ-t) and the temperature gradient data exceeding the threshold (G1'-G1 allow or G2'-G2 allow) in the verification data are input into the thermal inertia stress prediction model constructed in step 1.

[0115] The process of optimizing the signal generation is as follows: the model corrects the airflow regulation parameters through inverse calculation. The inverse calculation formula is:

[0116]

[0117] in, To optimize the amplitude of the adjustment signal, This is the amplitude of the previous basic airflow adjustment signal. , For correction factors (e.g.) , ), τ′ is the settling time of the temperature response curve, and τ is the thermal inertia hysteresis τ; For temperature gradients exceeding the threshold, The corresponding temperature gradient is preset with an allowable value; the signal direction is adjusted according to the deviation direction. When the temperature response curve deviation is positive, the signal direction is the same as the original signal, and when it is negative, it is the opposite; when the temperature gradient exceeds the threshold and is positive, the signal direction is to adjust the air volume to reduce the gradient.

[0118] The signal feedback and adjustment process is as follows: the optimization adjustment signal is fed back to the frequency converter of the fan, the frequency converter outputs frequency according to the amplitude of the signal, and then adjusts the fan speed to achieve air volume correction.

[0119] After completing the air volume maintenance or parameter correction operation, return to step 4, collect and verify the data again, judge the adjustment effect, and repeat steps 4-5 until the temperature difference of each temperature zone of the annealing furnace is stable within the target temperature range and the transverse and longitudinal temperature gradients of the glass belt are continuously within the preset allowable range, thus meeting the closed-loop conditions, stopping the cycle, and maintaining the current air volume output state.

[0120] This step achieves dynamic optimization of airflow through closed-loop control, which can continuously adapt to changes in operating conditions, ensure stable temperatures in each zone of the annealing furnace and meet the requirements for temperature gradient of the glass ribbon, thereby effectively controlling the risk of residual stress in the glass ribbon and improving the stability of glass annealing quality.

[0121] The embodiments described above are merely illustrative of specific implementations of the present invention, and while the descriptions are detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention.

Claims

1. A method for frequency conversion regulation of air volume and temperature control in an annealing kiln, characterized in that, Includes the following steps: Step 1: Collect basic operating data of the annealing furnace, and construct a thermal inertial stress prediction model based on the correlation mechanism between the heat transfer characteristics of the annealing furnace and the stress generation of glass. The process of constructing the thermal inertial stress prediction model is as follows: The lateral temperature gradient is... Longitudinal temperature gradient The glass strip thickness d, the real-time fan speed n, and the temperature T of each zone are used as model input parameters, and the thermal inertia hysteresis is included. Using the potential stress risk level R of the glass strip as the output parameter, a multivariate linear correlation model is constructed, and the model expression is: ； in These are the model coefficients. , For constant terms; Step 2: Using the aforementioned prediction model and combined with operational data, predict the thermal inertia hysteresis of the annealing furnace and the potential stress risk of the glass ribbon. Step 3: Based on the predicted thermal inertia hysteresis and temperature deviation in the basic operating data, a basic air volume adjustment signal is generated and transmitted to the frequency converter of the fan to complete the initial air volume adjustment. Step 4: After the initial airflow adjustment, collect verification data and judge the adjustment effect based on the prediction model and verification data; Step 5: Based on the judgment result, perform air volume maintenance or parameter correction operations, and repeat steps 4-5 until the temperature of each zone of the annealing furnace is stable and the stress risk of the glass strip is within the preset range.

2. The method for frequency conversion regulation and temperature control of air volume in an annealing kiln according to claim 1, characterized in that, The basic data collected for the operation of the annealing kiln includes: the spatial temperature of each zone of the annealing kiln, the temperature of multiple measuring points on the surface of the glass strip, the thickness of the glass strip, and the real-time operating parameters of the blower. The multiple measuring points on the surface of the glass strip are distributed along its width and the length of the kiln zone.

3. The method for frequency conversion regulation and temperature control of air volume in an annealing kiln according to claim 2, characterized in that, Step 1 involves constructing a thermal inertia stress prediction model, which includes: The transverse temperature gradient is calculated by measuring the temperature at different widths of the same cross section on the surface of the glass ribbon. The longitudinal temperature gradient is calculated by measuring the temperature at the same width location in different kiln zones.

4. The method for frequency conversion regulation and temperature control of air volume in an annealing kiln according to claim 3, characterized in that, The predicted thermal inertia hysteresis of the annealing furnace and the potential stress risk of the glass ribbon include: The collected glass strip thickness and the calculated transverse and longitudinal temperature gradients of the glass strip surface in each zone are input into the thermal inertia stress prediction model; the thermal inertia hysteresis of the annealing furnace under the current operating conditions and the potential stress risk level of the glass strip are output. The stress risk level is classified based on the correlation between the temperature gradient and the preset stress threshold.

5. The method for frequency conversion regulation and temperature control of air volume in an annealing kiln according to claim 2, characterized in that, The process of generating a basic airflow adjustment signal based on the predicted thermal inertia hysteresis and temperature deviation in the basic operating data, and transmitting it to the frequency converter of the fan to complete the initial airflow adjustment, includes: The collected spatial temperatures and glass surface temperatures of each zone are compared with the corresponding preset target temperatures to calculate the temperature deviation values ​​of each zone. Based on the temperature deviation value and the thermal inertia hysteresis predicted in step 2, a collaborative calculation is performed to generate a basic air volume adjustment signal that is adapted to the current operating conditions. The basic air volume adjustment signal is transmitted to the frequency converter control module of the fan in the corresponding area of ​​the annealing kiln, and the initial air volume output is achieved by adjusting the fan speed.

6. The method for frequency conversion regulation and temperature control of air volume in an annealing kiln according to claim 3, characterized in that, The verification data collected after the initial airflow adjustment included: Within a preset period after initial airflow adjustment, real-time temperature response curves of each zone and temperature data from multiple measuring points on the glass strip surface are collected. Repeat the calculation method of transverse and longitudinal temperature gradients in step 1, and update the current transverse and longitudinal temperature gradient data based on the temperature data of multiple measuring points on the newly collected glass strip surface. The real-time temperature response curves of each zone and the updated horizontal and vertical temperature gradient data were used as validation data.

7. The method for frequency conversion regulation and temperature control of air volume in an annealing kiln according to claim 6, characterized in that, Judging the adjustment effect based on the aforementioned prediction model and verification data includes: Analyze the matching degree between the real-time temperature response curves of each zone and the thermal inertia hysteresis predicted in step 2, and determine whether the temperature response meets the preset adjustment expectations. Determine whether the updated transverse and longitudinal temperature gradients on the glass strip surface are both within the corresponding preset allowable ranges. If the temperature response meets the preset adjustment expectation and both the lateral and longitudinal temperature gradients are within the preset allowable range, the adjustment is deemed effective; otherwise, the adjustment is deemed ineffective.

8. The method for frequency conversion regulation and temperature control of air volume in an annealing kiln according to claim 7, characterized in that, Maintaining the airflow in step 5 includes: If the adjustment is deemed effective, maintain the current fan speed and air volume output, and continuously monitor the temperature changes in each zone of the annealing furnace and the temperature gradient of the glass ribbon until the temperature in each zone is stable within the target temperature range and the transverse and longitudinal temperature gradients of the glass ribbon remain within the preset range.

Images